Driven by processes in the deep earth over millions of years, the East African Rift is slowly tearing the continent apart, producing earthquakes and volcanoes along its 2,400-mile track. A scientific team including Donna Shillington, James Gaherty and Cornelia Class of Columbia University’s Lamont-Doherty Earth Observatory is working in Malawi and Tanzania to understand the causes, the long-term evolution, and the real-time hazards.
The lakes along the Great African Rift Valley are among the largest fresh water lakes in the world. They lie in depressions created by slow stretching and thinning of the east African continent over millions of years. Many of the essential geological structures that enable the continent to tear and produce earthquakes are hidden within the Earth below these lakes. Lake Malawi (Nyasa) is the southernmost of these Great Rift Valley lakes and represents one of the youngest segments of the East African Rift System today. The lake is a whopping 550 km long and up to 70 km wide and surrounded by three countries : Mozambique to the southeast, Tanzania to the northeast, and Malawi to the west.
To image geologic structures and record earthquakes beneath northern Lake Malawi, our science team is undertaking a major “marine” seismic study as a part of the NSF-funded SEGMeNT (Study of Extension and maGmatism in Malawi aNd Tanzania) project. This part of the project involves generating sound waves using a towed array of “air guns” and recording the sound waves on a 1500-m-long cable filled with pressure sensors and an array of seismic stations deployed both onshore and on the lake bottom. The scientific and technical staff for this part of the project come from Lamont-Doherty Earth Observatory of Columbia University, Syracuse University, the Malawi Geological Survey Department, the Geological Survey of Tanzania, Aarhus University and Scripps Institution of Oceanography.
Marine seismic studies like ours are routinely done in the oceans using scientific equipment and research vessels outfitted specially for these purposes. Collecting comparable data in a great lake in Africa requires creative repurposing of available vessels and adaption of scientific equipment. To deploy and recover seismometers on the lake floor, Jim Gaherty and team used a small research vessel (R/V Ndunduma) operated by Malawi Fisheries Department. Deck space is limited, requiring efficient packing and multiple trips to deploy 34 seismometers in the lake with a boom normally used for dragging fishing nets. For the seismic imaging component, we transformed a large container ship (M/V Katundu) into a seismic research vessel. Containers were placed on the deck that house our scientific “lab,” a workshop for repairing science equipment, a storage space for extra gear and miscellaneous items, and an accommodation container with 8 bunks to sleep some of the science party. We have also added large spool for the seismic streamer, generators and compressors to drive the seismic sound source, and a large metal arm (termed “the ironing board”) for towing the seismic source. Using non-standard ships, equipment and data collection procedures requires a team with technical expertise and ingenuity, and happily we have that in spades.
We are now slowing steaming across beautiful Lake Malawi in the M/V Katundu acquiring fantastic data as we go …
Donna Shillington and Natalie Accardo, M/V Katundu, 22 March 2015
Ideally, seismic stations are sited in remote, quiet locations away from any possible cultural noise, especially people, who are very noisy (even if they are not New Yorkers). But other considerations besides peace and quiet are important for a good station, particularly security. As a result, we placed most of our stations in towns near schools, hospitals or town halls, where people could keep an eye on them.
We often attract crowds while installing our exotic seismic gear. Field work with an audience has pros and cons. It’s certainly somewhat distracting to labor and sweat under the sun, tinkering with wires and programming equipment with a big crowd in attendance. Some of the sites are in relatively tight spots, so the curious onlookers occupied much of our working space, making for very close quarters. Several days ago, we installed a station next to the village hall in Ndalisi as a small crowd looked on and an animated town meeting took place next door. Loud passionate speeches inside were matched by loud banging outside as we mounted a solar panel for our station on the roof.
But there are very big upsides. People from the villages where we deployed stations have provided an enormous amount of help with building our sites. We have also had abundant opportunities to tell people what we hope to learn about the active tectonic environment where they live. Continental rifting here gives rise to geohazards such as earthquakes and volcanoes. Because we have tried to locate many of our sites near schools, we particularly hope to communicate our science to students and teachers. At the Matema Beach High School, students peppered us with questions as we installed our gear. Their school is just a stone’s throw from the Livingstone Mountains, the surface expression of a major rift fault that has caused large earthquakes. But our seismic installations admittedly may not be entirely positive; today at Kifule Secondary School, students took a long math exam inside while we were making a racket outside. But hopefully the pros out weigh the cons… Even at Kifule, students burst out of classroom after the test all smiles, so apparently we were not too disruptive.
Driving around the Rungwe volcanic province in the southern East Africa Rift installing seismometers, we have the chance to observe first hand how geological processes in action create the most dramatic forms at Earth’s surface. Looming volcanoes flanked by cinder cones lie along the rift valley, often very close to rift faults. The Livingstone Mountains, the surface expression of a major fault system that bounds the rift to the east in this area, soar over 1.5 km over the valley below, including Lake Malawi (Nyasa).
The remarkable geological structures evident above ground motivate us to look deeper in the earth. We see volcanoes in particular places at the surface, but where are magmas located at depth below the volcanoes and the rift? Likewise, we see dramatic faults that are helping to thin and break the crust at the surface, but how do they relate to stretching of the entire crust and lithosphere beneath this part of the East Africa rift? And how are the magmas and faults related to one another? These are the core scientific questions motivating our study of the rift around northern Lake Malawi (Nyasa). We hope to use data collected during this program, including the 15 seismic stations that we are deploying now around the Rungwe province, to answer these big questions.
The last time we visited the southern part of the East Africa Rift, we were responding to an unusual series of earthquakes in December 2009 that shook northern Malawi. The faults responsible for these events had not produced any earthquakes historically, and thus caught everyone by surprise. The unexpected occurrence of earthquakes on these faults highlights our poor overall understanding of how the African continent is slowly stretching and breaking apart.
This time, we return to this part of the rift system as a part of a more comprehensive effort to understand the underpinnings of this continental rift using a spectrum of geological and geophysical tools and involving a big international team of scientists from the U.S., Tanzania and Malawi. In the coming three weeks, we plan to deploy ~15 seismometers in southwest Tanzania around the Rungwe volcanic province, the southernmost volcanism in the East Africa Rift system. These stations will record small local earthquakes associated with active shifting of faults and moving of magmas at depth. They will also record distant earthquakes that can be used to create images of structures beneath Earth’s surface and map the faults and magmas.
A new study in the journal Nature provides fresh insight into deep-earth processes driving apart huge sections of the earth’s crust. The process, called rifting, mostly takes place on seabeds, but can be seen in a few places on land—nowhere more visibly than in the Afar region of northern Ethiopia. (See the slideshow below.) Here, earthquakes and volcanoes have rent the surface over some 30 million years, forming part of Africa’s Great Rift Valley. What causes this, and does it resemble the processes on the seafloor, as many geologists think?
The study suggests that conventional ideas may be wrong. Past calculations done by scientists predict that the solid rock under the Afar should be stretching and thinning substantially as the continent tears apart; thus molten rock should not have far to travel to the surface. Led by David Ferguson, a postdoctoral researcher at Columbia University’s Lamont-Doherty Earth Observatory, researchers analyzed the chemical makeup of lava chunks they collected from the Afar. They showed that magmas actually came from quite deep–greater than 80 kilometers, or 45 miles, within the earth’s mantle–and formed under extraordinarily high temperatures, above 1,450 degrees C, or 2,600 F.
This implies that magmas are generated by a long-lasting plume of mantle heat. It also indicates that magma must make its way up through a surprisingly thick lid of solid rock, called the lithosphere. This idea has been supported by some seismic images of the Afar subsurface.
Rifting here is fairly slow—one or two centimeters a year, or 0.4 to 0.8 inches, and this may partly explain why so much solid rock persists. As the lithosphere is pulled apart, it does stretch, crack and thin. However, because the process in this region takes so long, the base of the lithosphere has time to cool down by losing heat to the colder rock above. This keeps the relatively cold, brittle lithosphere thicker than would be expected, and counteracts stretching. Sometimes, though, magma suddenly spurts long distances to the surface, and the earth visibly cracks and pulls apart during spectacular rifting events. That includes a series of events that started in 2005, and was closely observed by scientists.
Parts of the rift have already sunk below sea level. In the distant future–maybe 10 million years from now–the process will advance so far that the Red Sea will break through and flood the region. A new sea will open up, whether or not there is anyone around to name it.
In East Africa, earth’s crust is stretching and cracking, in a process called rifting. Here in the Afar region of northern Ethiopia, hundreds of faults and fissures have formed over time. (David Ferguson)
An important force driving the rifting is magma created beneath earth’s rocky outer shell, which has forced its way upward to push apart the crust. This eruption happened in the Afar in June 2009. (David Ferguson)
This crevice opened in a matter of hours, during a sequence of very large earthquakes in September 2005. It formed in response to magma being injected into the shallow crust, and is still emitting volcanic gases. This injection of magma was the largest event of its kind to be observed by scientists. (Lorraine Field)
Fresh lava erupted onto the desert floor preserves fragile surface textures, formed as the viscous molten rock cooled and hardened. Over time, these sharp features will erode away. (David Pyle)
A remote field site within the rift. Afar is one of the hottest and most sparsely populated regions on the planet. (David Pyle)
In a region that is vast, largely roadless and dominated by armed tribes, scientists depend on helicopters to get around, and on local people to act as guides and security guards. The climate necessitates large amounts of portable drinking water. (David Ferguson)
Lavas forming the rift surface cracked apart during an earthquake in 2005 to form this fault. The horizontal boundary between the light and dark area marks the pre-2005 ground surface, and shows that the area in the foreground dropped several meters during the quake. The geology of Afar provides many clues to the tectonic and magmatic process operating beneath our feet. (David Pyle)
In early May, Scott Nooner and I returned to Malawi to retrieve our seismic equipment and finally lay eyes on the data recorded over the last 4 months. Picking them up was vastly easier than putting them out. In contrast to the days studying out-dated maps and driving down dirt roads looking for sites, and hours of hard labor under the hot African sun digging holes and constructing vaults, recovery required only minutes at each site to shut down the equipment and safely stow it in plastic cases in the back of our rented truck. It took us about a day to recover all the equipment that we spent a week installing. Since we recovered the seismic equipment so quickly, we had time to collect new GPS data, too.
Although retrieving the seismic equipment proved easy, transporting numerous 50-lb boxes from one side of the world to the other is not trivial, as we discovered during the deployment. Our hasty departure in January prevented us from obtaining US customs documentation that would have simplified the export/import process, and our seismic equipment had to return from Malawi the way it came in – as checked luggage. We wrested ten ~50-lb pieces of baggage to the check-in counter at the Lilongwe airport, and handed over all of our remaining dollars plus a fistful of Malawi Kwacha for excess baggage fees. Checking in for each subsequent leg of the trip, we braced ourselves to part with more money. Even for the few pieces of equipment that we transported back to the US via a commercial shipper, we faced interesting challenges. The Karonga DHL office lacks a scale, so shipping agents and our Geological Survey colleagues made competing guesses as to the weight of our boxes and compromised on the average when charging us shipping fees.
Far and away the best part of recovering instruments is the chance to take a first look at the data, and our new dataset from Malawi did not disappoint. While sipping complementary wine on the long flight from Johannesburg to Atlanta, I perused the recordings from our seismometers. While (thankfully) there were no recurrences of the damaging events from December, to my delight I saw that we have recorded a remarkably persistent series of aftershocks. For our purposes, the more aftershocks, the merrier! We plan to determine the location of each aftershock to map out the structures below Earth’s surface that caused the large sequence of earthquakes in December. Stay tuned…..
While installing our seismic network in Malawi, we interacted with everyone from scientists to schoolteachers, and journalists to villagers. The opportunity to provide information and education to Malawians has been the most rewarding aspect of our effort. We trained local scientists and technicians on seismic equipment and data analysis, and educated the public on earthquakes and earthquake monitoring both in person and via media interviews. The Malawi Geological Survey Department (MGSD) prompted our visit by requesting assistance in monitoring aftershocks, and we hope that this temporary seismic deployment will empower them to obtain resources and training for a permanent seismic network.
Because we deployed our seismic stations near schools, clinics and other centers of village life, we met a wide spectrum of Malawians. Everyone we spoke with expressed interest in our undertaking and wanted to know more about the chindindindis (earthquakes in Tumbuka). In the village of Mpata, 5 miles west of Karonga, a crowd gathered around a laptop balanced on the hood of our 4×4 as Jim showed them aftershocks in newly downloaded data; the audience peppered him with pertinent questions about the East African Rift and earthquakes beneath Lake Malawi. Curious policeman looked on as I retrieved seismic records from a station positioned near a checkpoint ~10 miles north of Karonga, inquiring when and where the next earthquake would occur. Science teachers in Mlare helped us install a station near their school and received an impromptu lesson in plate tectonics and seismology.
Journalists from newspapers, radio stations and national TV programs also interviewed us during our visit, which allowed us to communicate with a larger audience about possible causes of the earthquakes and the benefits of monitoring them.
We worked side by side with scientists and technicians from the MGSD every day of our visit. They taught us local geology, local customs, and local language, and made our joint endeavor possible by facilitating contacts with national and regional officials. In return, we brought them seismic monitoring equipment, helped them deploy it, and taught them new techniques for analyzing the resulting data. Although the MGSD is charged with monitoring earthquakes within the Malawi rift valley, their efforts are severely hampered by paucity of data and lack of training. Only two seismic stations exist in Malawi (provided by Africa Array), and university-level courses in seismology are almost non-existent. The data and training of MGSD employees provided by our temporary deployment following the Karonga earthquakes will help mitigate these problems in the short term; we hope that this experience will equip the MGSD with the ammunition to argue for more national and international resources for seismic monitoring in Malawi over the longer term.
A rapid technical response to the damaging earthquakes in Malawi produces both humanitarian and scientific benefits, and we hoped that both scientific and international assistance agencies would support our effort. Our seismic field effort serves two purposes: (1) to provide badly needed seismic equipment and technical training to the Malawi Geological Survey department (MGSD); and (2) to obtain unique data from very close to the earthquake sources to develop a better scientific understanding of faulting in the East African Rift. Funding has proven difficult, however, and our experience suggests that a technical component to earthquake response often falls through the cracks of the broader relief effort.
The Malawi earthquake sequence spawned a modest international response by several organizations with complementary and overlapping goals. The US Agency for International Development (USAID), through the Office of Foreign Disaster Relief (OFDA), and international organizations (e.g., Red Cross) provided direct humanitarian response: food, water, shelter, and other necessities for the displaced people of Karonga. Two scientists from the US Geological Survey (USGS), with support from USAID, provided a post-earthquake assessment based on field observations of damage and faulting, which constituted the official US government technical response.
Our technical response parallels those efforts, and is typical for the US academic community; individual scientists with existing contacts in and working knowledge of the effected region provide seismological field equipment, analysis, and training. Responding to the earthquakes in a timely manner required an almost instantaneous commitment on our part. Within two days after the largest event, IRIS had mobilized instruments and the funding necessary to ship them to the field. Lamont-Doherty Earth Observatory (LDEO) and the Earth Institute (EI), both at Columbia University, promised to “backstop” our effort – in other words, cover our travel and field expenses while we sought external funding for our effort. Both have strong and long-standing commitments to mitigating earthquakes, hazards, and human suffering worldwide, including in East Africa and Malawi. The project would have immediately stalled without this support.
With the LDEO and EI backstop in hand, we sought external funds from the National Science Foundation (NSF) and USAID, highlighting the unique scientific and outreach opportunities offered by a rapid response to these earthquakes (read our proposal here). USAID characterized the activity as too scientific to be in their purview and declined to fund us. NSF acknowledged a modest scientific benefit, but they described the effort as primarily a humanitarian and outreach response. While NSF agreed to provide some support, the amount available for such short-turnaround projects (via the RAPID program) is very small – enough only to return and recover our instruments.
Technical responses such as this one provide scientific and humanitarian benefits alike and strongly complement the larger response effort. The breadth of the impact should increase their fundability – more bang for the buck. But because of the splintered nature of the US response and funding mechanisms, this breadth can be a detriment to obtaining funding – too scientific to be humanitarian, but too humanitarian to be scientific. In our case, we overcame this quandary only with the strong financial support of our home institution. How many technical response efforts never get off the ground because of this funding uncertainty?
The ideal spot for a seismic station is dry, quiet and safe from vandals and thieves. Seismometers record slight ground motions, allowing them to hear distant (and not so distant) earthquakes. But cars or even kids playing near a seismic station can produce ground vibrations that overwhelm the subtle sounds of earthquakes. Seismic stations include plenty of expensive, high-tech instruments that are worthless to the average person. But they also contain mundane items that can be useful, such as 12-volt batteries and insulated wiring, making theft a problem. And water is the enemy.
Malawi presented novel challenges for siting our stations. Our first priority was to find dry, secure locations to prevent damage and loss. As we drove into the Karonga region for the first time, our hearts sank; the epicentral region is low-lying and wet, small villages surrounded by rice paddies. Our arrival during the rainy season did not help.
But with a little hunting, we were able to find high and dry spots for most of our stations. We bumped along narrow village tracks in our rented 4×4, occasionally getting stuck on particularly muddy sections. Most of the dirt roads did not appear on our outdated maps, so we stopped regularly to ask for directions. When our Malawi colleagues explained that we were there to learn about the chindindindis (Tumbuka for earthquakes), they were eager to help!
In many parts of the world, safety and quiet can be achieved simultaneously simply by deploying stations in the middle of nowhere. This is not an option in densely populated Malawi, where one farming village abuts another. Main thoroughfares and small dirt roads alike were crowded with kids walking to school, villagers biking to town, and farmers grazing their goats and sheep. Instead, we sought out village police, teachers, and other officials for help finding safe spots. In some cases we hired guards to look after them.
We spent hours driving, inspecting sites and waiting to meet with officials. We normally skipped lunch, fueling ourselves instead on passion fruit-flavored Fantas and “puffs” (kids junk food akin to cheese doodles). But these efforts paid off – we found good sites for our equipment and started listening.
The magnitude 6.0 earthquake that struck Malawi on Saturday night, December 19, spurred us into action. We had been closely following the earthquakes there, but this one confirmed the unusual nature of the seismic sequence. It also happened to be the most destructive. Leonard Kalindekafe, director of Malawi’s Geological Survey, asked us to come and monitor the ongoing quakes. However, mobilizing the needed equipment over the holidays turned out to be a challenge.
Within two days, we located ten seismometers: eight from the Incorporated Research Institutions for Seismology (IRIS), and two from Cindy Ebinger, a seismologist at University of Rochester in upstate New York. Cindy offered to send us her instruments, which we would carry on the plane with us. IRIS planned to ship their instruments directly to Malawi. On Christmas Eve, as we headed home to our families, everything seemed to be in order. We purchased plane tickets for Dec. 30, and planned to celebrate New Year´s Eve in Lilongwe, Malawi’s capital, with a couple of Carlsberg Specials at the Diplomat Pub. From there, we would head north to the epicenter.
Not all of it happened as planned. Shipments were canceled. Boxes went missing. Flights were changed. Shipping the IRIS equipment directly to Malawi required two weeks in transit, minimum. We considered wild back-up plans: “Let’s truck everything from Johannesburg to Malawi!” Just as quickly we rejected them: Johannesburg is 1,000 miles away, and would have required four border crossings.
More or less everything that could go wrong, did. The delays and false-starts were particularly frustrating since the clock was ticking; the rate of aftershocks declines steadily following a major earthquake. Each day of delay meant less information about the origin of the big earthquakes.
After two days of arranging and rearranging, hair pulling and hand wringing, we departed New York on New Year’s Day with equipment for five seismic stations, all of it packed into our checked luggage. We even crammed two seismometers into our carry-on backpacks; they passed through security at JFK apparently unnoticed. Eighteen hours later our eight 50-lb bags arrived at the VIP customs lounge in Lilongwe. Leonard helped speed us and our equipment through customs. Within an hour, we were in a 4×4 speeding towards Karonga. Almost nothing went right prior to our departure from JFK; from that point forward, nothing really went wrong.